Interstitial cystitis (IC) is a chronic syndrome of the urinary bladder that primarily affects women. A broad clinical definition of IC includes any patient who complains of pelvic/perineal pain, urinary urgency, or voiding frequency in absence of any obvious cause such as infection or cancer. In 1993, the NIDDK established the Interstitial Cystitis Data Base (ICDB) as the first prospective longitudinal cohort study of patients with symptoms consistent with IC. The enrolled 637 patients diagnosed with IC and followed them for up to four years. Baseline measures included demographics, medical history, detailed symptom questionnaires, and urodynamic testing. Cytoscopy/hydrodistention and bladder biopsies were performed at the discretion of the treating physician. The proposed research involves targeted analyses using the ICDB to address clinically and scientifically relevant questions regarding the natural treated history of IC. This can provide insight into the epidemiology and etiology of IC, aid in the refinement of diagnostic criteria, and ideally identify groups to which specific treatment modalities may be targeted. There are three goals of the proposed research. The first is to identify subgroups of patients with IC characterized by constellations of symptoms, lifestyle factors, and/or biological markers. These analyses will involve both cross-sectional (baseline) and longitudinal comparisons between symptom severity and demographics, comorbidities, seleceted therapies, and cystometric measures. Detailed statisticall analyses of pathological features derived from bladder biopsies and their association with symptoms will be performed. The second goal is to conduct exploratory analyses of various outcome measures, both cross-sectionally and longitudinally, to provide guidelines for the design, conduct, and analysis of future clinical trials in IC. These analyses will include the evaluation of quality of life over time and the exploration of various definitions of response derived from longitudinal symptom changes. The third goal relates to the application and/or development of statistical methodology to aid in the identification of subgroups and evaluation of outcome measures. The methodology includes latent variable modeling of multiple outcome data, methods for tracking of symptoms in multiple domains over time, methods for assessing regression to the mean, and risk-set matching for evaluation of time-dependent treatment decisions. In addition, through this project, the ICDB will be maintained and made available to external investigators interested in using the data to explore specific hypotheses related to the natural history of IC.